This project deploys a yolo fastest model in the form of tflite on raspberry 3b+. The model is from another repository of mine called -Trash-Classification-Car

Overview

Deploy-yolo-fastest-tflite-on-raspberry

觉得有用的话可以顺手点个star嗷

这个项目将垃圾分类小车中的tflite模型移植到了树莓派3b+上面。
该项目主要是为了记录在树莓派部署yolo fastest tflite的流程

(之后有时间会尝试用C++部署来提升性能)

一些问题

1. 如何运行tflite文件?

关于如何在linux端运行tflite模型的问题,官方文档中已经给的非常清楚,详见tflite.API

2. yolo-fastest的解码问题?

由于yolo fastest的输出格式和其他版本的yolo不太一样,所以其yolo输出的解码模式和其他版本yolo不同,需要引起注意。若要部署的模型不是yolo fastest tflite而是其他yolo,该项目可能不能直接适用, 但根据能力进行修改即可。

3.yolo-fastest的源码:Yolo-Fastest

现在yolo fastest的作者推出了V2版本,性能更好。该项目采用的是V1.

4.关于如何在windows上训练yolo fastest模型,详见本人另一个仓库:Yolo-Fastest-on-Windows

模型实机效果

image
该项目在树莓派3b+上可以跑到平均25帧每秒。
image

小车实机效果

image

项目内容

本项目包含两个文件夹,detect-camera-streamdetect-single-img
两个文件夹中结构相同,模型文件存在两个文件夹下的tflite/Sample_TFlite_model中,主程序写在TFLite_detection_stream.py和TFLite_detection_img.py里,yolo相关的函数写在yolo_layer.py中。

detect-camera-stream文件可以在树莓派3b+连接USB摄像头的情况下,实时的对视频流进行目标检测。
detect-single-img文件可以对tflite/下的4.jpg图片,即单独一张图片进行检测。

关于运行的命令,存放在instruction.txt之中。

如何直接运行该项目:

  1. 确保树莓派上有python3.7解释器。

  2. 安装virtualenv:

python3 -m venv tflite-env 

  3. 下载该项目所有文件。
  4. 进入tflite文件夹,进入虚拟python环境:

source tflite-env/bin/activate
bash get_pi_requirements.sh                  :若上一步提示缺少环境则执行这一行

  5. 在tflite文件夹下,运行instruction.txt中的指令:

python3 TFLite_detection_image.py
python3 TFLite_detection_stream.py

从零部署流程

  以detect-camera-stream为例。

  1. 创建虚拟python环境:

  创建一个tflite文件夹,创建虚拟环境:

cd tflite                                     :进入tflite
sudo pip3 install virtualenv                  :创建虚拟环境需要的工具
python3 -m venv tflite-env                    :创建虚拟环境,虚拟环境储存在tflite/tflite-env中
source tflite-env/bin/activate                :进入虚拟环境,每次推出terminal后都要执行此命令以进入虚拟环境

  2. 安装包和依赖:

  在进入虚拟环境后,提取出该项目中的get_pi_requirements.sh,放在tflite文件夹下:

bash get_pi_requirements.sh                   :下载包和依赖

  此时可通过以下代码来测试cv2模块是否安装好(opencv-python模块经常抽风):

python3
import cv2

  此时若没有报错则说明opencv-python安装成功,但经常出现以下错误:

ImportError: libjasper.so.1: cannot open shared object file: No such file or directory

  这个报错说明少安装了依赖,执行以下命令即可:(我是这样解决的,若解决不了请百度)

sudo apt-get install libjasper-dev

  3. 在tflite文件夹下创建Sample_TFlite_model文件夹,其中存放训练好的tflite模型。

  4. 运行模型 在tflite文件夹下,运行:

python3 TFLite_detection_stream.py

  即可看到效果

注意:若是自己的训练的模型而不是该项目里的,需要到TFLite_detection_stream.py中修改图片分辨率等参数。

由于树莓派要和小车通信,因此这里在记录一下在树莓派用AMA0实现串口通信的过程。

首先安装gedit编辑器,比vim好用一些:

sudo apt-get install gedit

然后禁用串口启动,开启串口硬件:

sudo raspi-config
interfacing options --> would you like a login shell to be accessible  over serial? --> No
                    --> would you like the serial port hardware to be enabled? --> Yes

由于蓝牙和AMA0使用的是同一个GPIO,将ttyAMA0和ttyS0的映射调换:

sudo gedit /boot/config.txt
在最后一行添加:dtoverlay=pi3-miniuart-bt
sudo reboot

因为控制台使用串口和通信串口只能存在一个,所以要禁用控制台来使用串口:

sudo systemctl stop [email protected]
sudo systemctl disable [email protected]

然后删除serial0相关:

sudo gedit /boot/cmdline.txt
删除 console=serial0,115200 ,没有就不管
sudo reboot

至此串口设置就完了,因为树莓派的python3解释器自带serial库,但我们之前创建的虚拟环境没有,所以要在虚拟环境再次安装:

sudo pip3 install pyserial
sudo pip3 install serial

可以通过以下代码来控制串口:

import serial
ser = serial.Serial('/dev/ttyAMA0',115200)      # 获取串口
if(ser.isOpen):
  ser.write(b'123')                             # 出现编码问题可以尝试加上 .encode()

通过GPIO来控制识别的开始和结束

这一部分的文件在该仓库的GPIO文件夹中可找到。

由于通过ssh连接树莓派比较复杂,且每次运行程序都需要电脑在手边,因此若能通过树莓派自身来控制程序的跑与结束,是最方便不过的了。
因此,我选择用一个按键开关来控制
image
这是一个双刀双掷开关,这里只用其中两个引脚。
  1. 写一个脚本来实现启动py文件:
  在/home/pi目录下编写charlie.sh文件:

cd /home/pi/Desktop/demo1/tflite
source tflite-env/bin/activate
python3 TFLite_detection_stream.py

  此时通过命令行输入bash /home/pi/charlie.sh即可运行py文件。
  2. 连线:
  将树莓派的3,5引脚连到开关的一段,GND连接到另一端。
  这样,在初始化时将3,5拉高。当开关按下时,3被拉低,可以此作为启动程序的标志。
  当开关被松开后,5被拉高,可以此作为退出程序的标志。
  3. 编写GPIO.py:
  首先在虚拟环境中安装RPi库:

pip3 install RPi.GP

  在/home/pi/Desktop/demo1/tflite的目录下编写GPIO.py,使当引脚3被拉低后运行charlie.sh:

import time
import RPi.GPIO as GPIO
import os

run_yolo_cmd = 'bash /home/pi/charlie.sh'

GPIO.setmode(GPIO.BOARD)
GPIO.setup(3, GPIO.IN, pull_up_down=GPIO.PUD_UP)

while(True):
    while(True):
        x = GPIO.input(3)
        if(x == 0):
            break

    print('pressed')
    time.sleep(1)
    os.system(run_yolo_cmd)
    

    while(True):
        x = GPIO.input(3)
        if (x == 1):
            break
    print('not pressed')
    time.sleep(1)

  注意,time.sleep(1)是必要的,因为按键在按下和松开时,电压是不稳定的,延时可以消抖。
  4. 修改TFLite_detection_stream.py:
  修改TFLite_detection_stream.py以使得其拥有检测到引脚5升高后自动结束运行的功能:

在TFLite_detection_stream中作如下添加:
import RPi.GPIO as GPIO

GPIO.setmode(GPIO.BOARD)
GPIO.setup(5, GPIO.IN, pull_up_down=GPIO.PUD_UP)

在进行模型推理的大循环中添加:
x = GPIO.input(5)
if(x == 1):
    print('camera exit')
    sys.exit(0)

  5. 实现开机自动运行GPIO.py

sudo gedit /etc/rc.local
在exit 0的上一行添加:
python3 /home/pi/Desktop/demo1/tflite/GPIO.py &
(&符号使得其一直在后台运行)

至此,开机后会自动运行GPIO.py,GPIO.py会不停检测引脚3。当按下引脚3后,GPIO.py会调用charlie.py来运行TFLite_detection_stream.py。TFLite_detection_stream.py会检测引脚5,当按键松开后,TFLite_detection_stream.py会自动退出。这是一个循环。再按下会再启动......

Cross-platform CLI tool to generate your Github profile's stats and summary.

ghs Cross-platform CLI tool to generate your Github profile's stats and summary. Preview Hop on to examples for other usecases. Jump to: Installation

HackerRank 134 Dec 20, 2022
Distributed Asynchronous Hyperparameter Optimization better than HyperOpt.

UltraOpt : Distributed Asynchronous Hyperparameter Optimization better than HyperOpt. UltraOpt is a simple and efficient library to minimize expensive

98 Aug 16, 2022
Not Suitable for Work (NSFW) classification using deep neural network Caffe models.

Open nsfw model This repo contains code for running Not Suitable for Work (NSFW) classification deep neural network Caffe models. Please refer our blo

Yahoo 5.6k Jan 05, 2023
PenguinSpeciesPredictionML - Basic model to predict Penguin species based on beak size and sex.

Penguin Species Prediction (ML) 🐧 👨🏽‍💻 What? 💻 This project is a basic model using sklearn methods to predict Penguin species based on beak size

Tucker Paron 0 Jan 08, 2022
An implementation of a sequence to sequence neural network using an encoder-decoder

Keras implementation of a sequence to sequence model for time series prediction using an encoder-decoder architecture. I created this post to share a

Luke Tonin 195 Dec 17, 2022
unet-family: Ultimate version

unet-family: Ultimate version 基于之前my-unet代码,我整理出来了这一份终极版本unet-family,方便其他人阅读。 相比于之前的my-unet代码,代码分类更加规范,有条理 对于clone下来的代码不需要修改各种复杂繁琐的路径问题,直接就可以运行。 并且代码有

2 Sep 19, 2022
一个目标检测的通用框架(不需要cuda编译),支持Yolo全系列(v2~v5)、EfficientDet、RetinaNet、Cascade-RCNN等SOTA网络。

一个目标检测的通用框架(不需要cuda编译),支持Yolo全系列(v2~v5)、EfficientDet、RetinaNet、Cascade-RCNN等SOTA网络。

Haoyu Xu 203 Jan 03, 2023
Repository of the paper Compressing Sensor Data for Remote Assistance of Autonomous Vehicles using Deep Generative Models at ML4AD @ NeurIPS 2021.

Compressing Sensor Data for Remote Assistance of Autonomous Vehicles using Deep Generative Models Code and supplementary materials Repository of the p

Daniel Bogdoll 4 Jul 13, 2022
Pytorch Implementation of PointNet and PointNet++++

Pytorch Implementation of PointNet and PointNet++ This repo is implementation for PointNet and PointNet++ in pytorch. Update 2021/03/27: (1) Release p

Luigi Ariano 1 Nov 11, 2021
We will release the code of "ConTNet: Why not use convolution and transformer at the same time?" in this repo

ConTNet Introduction ConTNet (Convlution-Tranformer Network) is proposed mainly in response to the following two issues: (1) ConvNets lack a large rec

93 Nov 08, 2022
Anomaly detection analysis and labeling tool, specifically for multiple time series (one time series per category)

taganomaly Anomaly detection labeling tool, specifically for multiple time series (one time series per category). Taganomaly is a tool for creating la

Microsoft 272 Dec 17, 2022
GANTheftAuto is a fork of the Nvidia's GameGAN

Description GANTheftAuto is a fork of the Nvidia's GameGAN, which is research focused on emulating dynamic game environments. The early research done

Harrison 801 Dec 27, 2022
PyTorch implementation for "Sharpness-aware Quantization for Deep Neural Networks".

Sharpness-aware Quantization for Deep Neural Networks This is the official repository for our paper: Sharpness-aware Quantization for Deep Neural Netw

Zhuang AI Group 30 Dec 19, 2022
Code for MSc Quantitative Finance Dissertation

MSc Dissertation Code ReadMe Sector Volatility Prediction Performance Using GARCH Models and Artificial Neural Networks Curtis Nybo MSc Quantitative F

2 Dec 01, 2022
Finite difference solution of 2D Poisson equation. Can handle Dirichlet, Neumann and mixed boundary conditions.

Poisson-solver-2D Finite difference solution of 2D Poisson equation Current version can handle Dirichlet, Neumann, and mixed (combination of Dirichlet

Mohammad Asif Zaman 34 Dec 23, 2022
Large-scale language modeling tutorials with PyTorch

Large-scale language modeling tutorials with PyTorch 안녕하세요. 저는 TUNiB에서 머신러닝 엔지니어로 근무 중인 고현웅입니다. 이 자료는 대규모 언어모델 개발에 필요한 여러가지 기술들을 소개드리기 위해 마련하였으며 기본적으로

TUNiB 172 Dec 29, 2022
Fine-grained Post-training for Improving Retrieval-based Dialogue Systems - NAACL 2021

Fine-grained Post-training for Multi-turn Response Selection Implements the model described in the following paper Fine-grained Post-training for Impr

Janghoon Han 83 Dec 20, 2022
Implementation of Hire-MLP: Vision MLP via Hierarchical Rearrangement and An Image Patch is a Wave: Phase-Aware Vision MLP.

Hire-Wave-MLP.pytorch Implementation of Hire-MLP: Vision MLP via Hierarchical Rearrangement and An Image Patch is a Wave: Phase-Aware Vision MLP Resul

Nevermore 29 Oct 28, 2022
Contextual Attention Network: Transformer Meets U-Net

Contextual Attention Network: Transformer Meets U-Net Contexual attention network for medical image segmentation with state of the art results on skin

Reza Azad 67 Nov 28, 2022
A self-supervised 3D representation learning framework named viewpoint bottleneck.

Pointly-supervised 3D Scene Parsing with Viewpoint Bottleneck Paper Created by Liyi Luo, Beiwen Tian, Hao Zhao and Guyue Zhou from Institute for AI In

63 Aug 11, 2022